From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753259AbbAXNjo (ORCPT ); Sat, 24 Jan 2015 08:39:44 -0500 Received: from mx1.redhat.com ([209.132.183.28]:58016 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752081AbbAXNjj (ORCPT ); Sat, 24 Jan 2015 08:39:39 -0500 Message-ID: <54C3A08D.3070301@redhat.com> Date: Sat, 24 Jan 2015 08:39:25 -0500 From: Rik van Riel User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.3.0 MIME-Version: 1.0 To: "H. Peter Anvin" , Suresh Siddha CC: Andy Lutomirski , Thomas Gleixner , Ingo Molnar , Fenghua Yu , the arch/x86 maintainers , Oleg Nesterov , linux-kernel Subject: Re: question about save_xstate_sig() - WHY DOES THIS WORK? References: <54C2A245.4010307@redhat.com> <54C2B80B.8050501@zytor.com> In-Reply-To: <54C2B80B.8050501@zytor.com> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 01/23/2015 04:07 PM, H. Peter Anvin wrote: > On 01/23/2015 11:34 AM, Rik van Riel wrote: >> While working on a patch series to defer FPU state loading until >> kernel -> user space transition, and be more lazy with FPU state >> while in the kernel, I came across this code in >> save_xstate_sig(). >> >> Not only is this broken with my new code, but it looks like it >> may be broken with the current code, too... >> >> Specifically, save_user_xstate() may page fault and sleep. After >> returning from the page fault, there is no guarantee that the FPU >> state will be restored into the CPU, when the system is not >> running with eager fpu mode. >> >> In that case, what prevents us from saving random FPU register >> state to the user's stack frame? Potentially state containing >> data from other programs... >> > > If the FPU state is not current, we'll have CR0.TS = 1 and the > XSAVE will cause an #NM exception, which will cause the FPU state > to be swapped in. OK, so the code works as it is right now, but I still need to have the patch in my "defer FPU state loading and CR0.TS switching to kernel -> user space transition" patch series. Thanks for explaining why the current code is safe, Peter. - -- All rights reversed -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEcBAEBAgAGBQJUw6CNAAoJEM553pKExN6DIkIH/jYpLRfVSbx+JBuFtOibA6jk wwWnwNdPRUiWf0v0lSNX4phk9f7VCy8WKbrFtC0IYdGLL0zCY94dN7us4hNYo9fE btdb/qsvqq8w2esPuUczAgaxwipYVliDUDOHSXTRbPS2upBuLHl/CVIVb1A+HVC7 gB6GF4ZLa2P1ygcjuWDNDNtx04QWTi7K80eWtWAQ7U4OK9Y4fSgR3hOIJS79M/aJ P9YGaTyo0zDQcKx1rFXTjl9A0slCFZ+QkAeCwLZRkXc1K9n50GANuYeg53/ATYey 7x0FjZEBD2ipX60FqZ+mYAiA7rqw4RYKgAttszy9b9bgKY/fzHy87z2YS8Umplg= =55wt -----END PGP SIGNATURE-----